In the examples given in the paper, there is an alternative way to describe what is happening; it may or may not be informative, and it may or may not hold up if a larger range of examples is used; in particular, the authors' own political positions may have influenced their choices of examples.

With those caveats out of the way: in each example in the paper, what happens is that both more science knowledge and more science curiosity tend to make people more like the liberal Democrats rather than conservative Republicans, except that science knowledge pushes conservative Republicans in the other direction.

Obviously this way of describing things is going to be more congenial to progressives than to conservatives. It would be interesting to see whether it remains a good description over a wide range of questions (in which case it might indicate that "reality has a liberal bias", or that a progressive conspiracy has taken over the scientific establishment, or something), or whether there's a broad range of questions where it goes the opposite way -- more science knowledge and curiosity push you rightwards, except that knowledge alone pushes lefties further left. (Possible places to look for such questions: areas where technocratic and environmentalist tendencies diverge, like support for nuclear power and genetically modified foods; areas where there's at least a case to be made that scientific research produces "politically incorrect" conclusions, like the alleged relationship between race and IQ. Any more?)

(If it turns out that the latter is what happens, an obvious guess is that the authors happen to be towards the left/progressive/liberal end of the scale, and picked examples of politically charged scientific questions that were salient to them.)

Interesting part of the article, for me:

“Partisans with weak math skills were 25 percentage points likelier to get the answer right when it fit their ideology,” Ezra Klein explained in a profile of Kahan’s work. “Partisans with strong math skills were 45 percentage points likelier to get the answer right when it fit their ideology. The smarter the person is, the dumber politics can make them.”

The title of this article is misleading (and the subtitle is just wrong). The research being summarized is a new paper "Science Curiosity and Political Information Processing" by Dan Kahan and others. They report that while partisanship leads to politically motivated reasoning, greater "science curiosity" tended to negate this. Subjects with higher "science intelligence" used this skill to engage more effectively in politically motivated reasoning, so that Democrats and Republicans views diverged more strongly with increased "science intelligence". However, Democrats and Republicans views converged slightly, while still being in disagreement, with increased "science curiosity". Science curiosity "reflects the motivation to seek out and consume scientific information for personal pleasure."

“We observed this kind of strange thing about these people who are high in science curiosity,” he says. The more scientifically curious a person, the less likely she was to show partisan bias in answering questions. “They seem to be moving in lockstep rather than polarizing as they became more science-curious.”